US2023376758A1PendingUtilityA1

Multi-modality root cause localization engine

Assignee: NEC LAB AMERICA INCPriority: May 20, 2022Filed: Apr 19, 2023Published: Nov 23, 2023
Est. expiryMay 20, 2042(~15.8 yrs left)· nominal 20-yr term from priority
G06N 3/04G06N 20/00G06N 3/045G06N 3/0455G06N 3/08G06N 3/0442G06N 7/01G06F 11/0709G06F 11/0769G06F 11/079G06F 21/552G06F 21/577G06F 2221/034G06F 2221/2101G06N 3/088G06N 3/047G06N 3/048G06N 20/10G06N 3/042G06N 5/041
75
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

A method for employing root cause analysis is presented. The method includes embedding, by an embedding layer, a sequence of events into a low-dimension space, employing a feature extractor and representation learner to convert log data from the sequence of events to time series data, the feature extractor including an auto-encoder model and a language model, and detecting root causes of failure or fault activities from the time series data.

Claims

exact text as granted — not AI-modified
What is claimed is: 
     
         1 . A method for employing root cause analysis, the method comprising:
 embedding, by an embedding layer, a sequence of events into a low-dimension space;   employing a feature extractor and representation learner to convert log data from the sequence of events to time series data, the feature extractor including an auto-encoder model and a language model; and   detecting root causes of failure or fault activities from the time series data.   
     
     
         2 . The method of  claim 1 , wherein the auto-encoder model includes a long short-term memory (LSTM) encoder network and an LSTM decoder network. 
     
     
         3 . The method of  claim 2 , wherein the LSTM encoder network learns a representation of a whole sequence as follows:
     f   t =σ g ( W   f   x   t   i   +U   f   h   t−1   +b   f )
       i   t =σ g ( W   i   x   t   i   +U   i   h   t−1   +b   i )
       o   t   =o   g ( W   o   x   t   i   +U   o   h   t−1   +b   o )       c   t =tanh( W   c   x   t   +U   c   h   t−1   +b   c )       c   t   =f   t   ⊙c   t−1   +i   t   +c   t          h   t   =o   t   ⊙c   t ,   where x t  is an input embedding of the t th  element in a training set S i , f t , i t , o t  are a forget gate, an input gate, and an output gate, respectively, W * , U * , and b * (*∈{f, i, o, c}) are all trainable parameters of the LSTM, and h N     i    is a final hidden state obtained by the LSTM.   
     
     
         4 . The method of  claim 2 , wherein the LSTM decoder network reconstructs the whole sequence as follows:
     h   t   i =LSTM( h   t−1   ,x   t−1   i )       p   t   i =Softmax(ReLU( W   p   h   t   i   +b   p ))       e   t   i =OneHot(argmax( p   t   i )       x   t   i   =E   T   ·e   t   i ,   where p t   i ∈R |E|  is a probability distribution over all possible events, W p  and b p  are trainable parameters, argmax is a function to obtain an index of largest entry of p t   i , Softmax normalizes the probability distribution, ReLU is an activation function defined as ReLU(x)=max(0,x),e t   i  is a predicted event at step t, and h N     i    represents hidden states.   
     
     
         5 . The method of  claim 2 , wherein parameters of the LSTM encoder network and the LSTM decoder network are optimized by an objective function defined as: 
       
         
           
             
               
                 
                   L 
                   
                     A 
                     ⁢ 
                     E 
                   
                 
                 = 
                 
                   
                     ∑ 
                     
                       i 
                       = 
                       1 
                     
                     N 
                   
                   
                     
                       ∑ 
                       
                         j 
                         = 
                         1 
                       
                       
                         N 
                         i 
                       
                     
                     
                       - 
                       
                         log 
                         ⁡ 
                         ( 
                         
                           
                             e 
                             j 
                             iT 
                           
                           ⁢ 
                           
                             p 
                             j 
                             i 
                           
                         
                         ) 
                       
                     
                   
                 
               
               , 
             
           
         
         where e j   i     T    is a predicted event and p j   i  is a probability distribution over all possible events. 
       
     
     
         6 . The method of  claim 1 , wherein the language model is trained to predict a next event given previous events in a categorical sequence. 
     
     
         7 . The method of  claim 6 , wherein the next event is predicted as:
     h   t   i =LSTM(( x   1   i   ,x   2   i   , . . . ,x   t   i ))       p   t+1   i =Softmax(ReLU( W   l   h   t   i   +b   l ))       e   t+1   i =OneHot(argmax( p   t+1   i )),   where p t+1   i  a probability distribution over all possible event, e t+1   i  is a one-hot representation the predicted next event, Softmax normalizes the probability distribution, ReLU is an activation function defined as ReLU(x)=max(0,x), argmax is a function to obtain an index of largest entry of p t   i , W l  and b l  are trainable parameters, and h t   i  represent hidden states.   
     
     
         8 . A non-transitory computer-readable storage medium comprising a computer-readable program for employing root cause analysis, wherein the computer-readable program when executed on a computer causes the computer to perform the steps of:
 embedding, by an embedding layer, a sequence of events into a low-dimension space;   employing a feature extractor and representation learner to convert log data from the sequence of events to time series data, the feature extractor including an auto-encoder model and a language model; and   detecting root causes of failure or fault activities from the time series data.   
     
     
         9 . The non-transitory computer-readable storage medium of  claim 8 , wherein the auto-encoder model includes a long short-term memory (LSTM) encoder network and an LSTM decoder network. 
     
     
         10 . The non-transitory computer-readable storage medium of  claim 9 , wherein the LSTM encoder network learns a representation of a whole sequence as follows:
     f   t =σ g ( W   f   x   t   i   +U   f   h   t−1   +b   f )
       i   t =σ g ( W   i   x   t   i   +U   i   h   t−1   +b   i )
       o   t =σ g ( W   o   x   t   i   +U   o   h   t−1   +b   o )
       c   t =tanh( W   c   x   t   +U   c   h   t−1   +b   c )       c   t   =f   t   ⊙c   t−1   +i   t   +c   t          h   t   =o   t   ⊙c   t ,   where x t  is an input embedding of the t th  element in a training set S i , f t , i t , o t  are a forget gate, an input gate, and an output gate, respectively, W * , U * , and b *  (*∈{f,i,o,c}) are all trainable parameters of the LSTM, and h N     i    is a final hidden state obtained by the LSTM.   
     
     
         11 . The non-transitory computer-readable storage medium of  claim 9 , wherein the LSTM decoder network reconstructs the whole sequence as follows:
     h   t   i =LSTM( h   t−1   ,x   t−1   i )       p   t   i =Softmax(ReLU( W   p   h   t   i   +b   p ))       e   t   i =OneHot(argmax( p   t   i )       x   t   i   =E   T   ·e   t   i ,   where p t   i ∈R |E|  is a probability distribution over all possible events, W p  and b p  are trainable parameters, argmax is a function to obtain an index of largest entry of p t   i , Softmax normalizes the probability distribution, ReLU is an activation function defined as ReLU(x)=max(0,x),e t   i  is a predicted event at step t, and h N     i    represents hidden states.   
     
     
         12 . The non-transitory computer-readable storage medium of  claim 9 , wherein parameters of the LSTM encoder network and the LSTM decoder network are optimized by an objective function defined as: 
       
         
           
             
               
                 
                   L 
                   
                     A 
                     ⁢ 
                     E 
                   
                 
                 = 
                 
                   
                     ∑ 
                     
                       i 
                       = 
                       1 
                     
                     N 
                   
                   
                     
                       ∑ 
                       
                         j 
                         = 
                         1 
                       
                       
                         N 
                         i 
                       
                     
                     
                       - 
                       
                         log 
                         ⁡ 
                         ( 
                         
                           
                             e 
                             j 
                             iT 
                           
                           ⁢ 
                           
                             p 
                             j 
                             i 
                           
                         
                         ) 
                       
                     
                   
                 
               
               , 
             
           
         
         where e j   i     T    is a predicted event and p j   i  is a probability distribution over all possible events. 
       
     
     
         13 . The non-transitory computer-readable storage medium of  claim 8 , wherein the language model is trained to predict a next event given previous events in a categorical sequence. 
     
     
         14 . The non-transitory computer-readable storage medium of  claim 13 , wherein the next event is predicted as:
     h   t   i =LSTM(( x   1   i   ,x   2   i   , . . . ,x   t   i ))       p   t+1   i =Softmax(ReLU( W   l   h   t   i   +b   l ))       e   t+1   i =OneHot(argmax( p   t+1   i ))   where p t+1   i  a probability distribution over all possible event, e t+1   i  is a one-hot representation the predicted next event, Softmax normalizes the probability distribution, ReLU is an activation function defined as ReLU(x)=max(0,x), argmax is a function to obtain an index of largest entry of p t   i , W l  and b l  are trainable parameters, and h t   i  represent hidden states.   
     
     
         15 . A system for employing root cause analysis, the system comprising:
 a processor; and   a memory that stores a computer program, which, when executed by the processor, causes the processor to:
 embed, by an embedding layer, a sequence of events into a low-dimension space; 
 employ a feature extractor and representation learner to convert log data from the sequence of events to time series data, the feature extractor including an auto-encoder model and a language model; and 
 detect root causes of failure or fault activities from the time series data. 
   
     
     
         16 . The system of  claim 15 , wherein the auto-encoder model includes a long short-term memory (LSTM) encoder network and an LSTM decoder network. 
     
     
         17 . The system of  claim 16 , wherein the LSTM encoder network learns a representation of a whole sequence as follows:
     f   t =σ g ( W   f   x   t   i   +U   f   h   t−1   +b   f )
       i   t =σ g ( W   i   x   t   i   +U   i   h   t−1   +b   i )
       o   t =σ g ( W   o   x   t   i   +U   o   h   t−1   +b   o )
       c   t =tanh( W   c   x   t   +U   c   h   t−1   +b   c )       c   t   =f   t   ⊙c   t−1   +i   t   +c   t          h   t   =o   t   ⊙c   t ,   where x t  is an input embedding of the t th  element in a training set S i , f t , i t , o t  are a forget gate, an input gate, and an output gate, respectively, W * , U * , and b *  (*∈{f,i,o,c}) are all trainable parameters of the LSTM, and h N     i    is a final hidden state obtained by the LSTM.   
     
     
         18 . The system of  claim 16 , wherein the LSTM decoder network reconstructs the whole sequence as follows:
     h   t   i =LSTM( h   t−1   ,x   t−1   i )       p   t   i =Softmax(ReLU( W   p   h   t   i   +b   p ))       e   t   i =OneHot(argmax( p   t   i )       x   t   i   =E   T   ·e   t   i ,   where p t   i ∈R |E|  is a probability distribution over all possible events, W p  and b p  are trainable parameters, argmax is a function to obtain an index of largest entry of p Softmax normalizes the probability distribution, ReLU is an activation function defined as ReLU(x)=max(0,x),e t   i  is a predicted event at step t, and h N     i    represents hidden states.   
     
     
         19 . The system of  claim 16 , wherein parameters of the LSTM encoder network and the LSTM decoder network are optimized by an objective function defined as: 
       
         
           
             
               
                 
                   L 
                   
                     A 
                     ⁢ 
                     E 
                   
                 
                 = 
                 
                   
                     ∑ 
                     
                       i 
                       = 
                       1 
                     
                     N 
                   
                   
                     
                       ∑ 
                       
                         j 
                         = 
                         1 
                       
                       
                         N 
                         i 
                       
                     
                     
                       - 
                       
                         log 
                         ⁡ 
                         ( 
                         
                           
                             e 
                             j 
                             iT 
                           
                           ⁢ 
                           
                             p 
                             j 
                             i 
                           
                         
                         ) 
                       
                     
                   
                 
               
               , 
             
           
         
         where e j   i     T    is a predicted event and p j   i  is a probability distribution over all possible events. 
       
     
     
         20 . The system of  claim 15 , wherein the language model is trained to predict a next event given previous events in a categorical sequence and wherein the next event is predicted as:
     h   t   i =LSTM(( x   1   i   ,x   2   i   , . . . ,x   t   i ))       p   t+1   i =Softmax(ReLU( W   l   h   t   i   +b   l ))       e   t+1   i =OneHot(argmax( p   t+1   i )),   where p t+1   i  a probability distribution over all possible event, e t+1   i  is a one-hot representation the predicted next event, Softmax normalizes the probability distribution, ReLU is an activation function defined as ReLU(x)=max(0,x), argmax is a function to obtain an index of largest entry of p t   i , W l  and b l  are trainable parameters, and h t   i  represent hidden states.

Join the waitlist — get patent alerts

Track US2023376758A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.